WO2009133123A1 - Verfahren und vorrichtung zum erkennen eines zustandes einer zu untersuchenden geräuscherzeugenden maschine - Google Patents
Verfahren und vorrichtung zum erkennen eines zustandes einer zu untersuchenden geräuscherzeugenden maschine Download PDFInfo
- Publication number
- WO2009133123A1 WO2009133123A1 PCT/EP2009/055165 EP2009055165W WO2009133123A1 WO 2009133123 A1 WO2009133123 A1 WO 2009133123A1 EP 2009055165 W EP2009055165 W EP 2009055165W WO 2009133123 A1 WO2009133123 A1 WO 2009133123A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- sound
- examined
- acoustic
- classification model
- generating
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
- G01M15/04—Testing internal-combustion engines
- G01M15/12—Testing internal-combustion engines by monitoring vibrations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N99/00—Subject matter not provided for in other groups of this subclass
Definitions
- the invention relates to a method and a device for detecting a state of a noise-generating machine to be examined, which emits body or airborne sound.
- Machines perform movements in the course of which vibration signals, in particular acoustic vibration signals, are generated.
- the vibration signals generated by a machine, a system or an electromechanical device allow a conclusion to their respective current state, which can change, for example due to wear and tear. For example, after several years of use, a machine or machine will produce different vibration signals due to signs of wear than directly after manufacture.
- chemical equipment can also cause noise signals, for example due to gas bubbles in containers or pipes. Due to the aging or wear of equipment, machinery or equipment, the generated vibration signals, in particular the acoustic emissions change.
- the vibration signals generated by a machine, a system or a device depend on the type of construction, in particular the acoustic noise signals.
- plants or devices have different product types or -modeile within a product group.
- a company can produce water or heating pumps in different variants whose generated noise signal is different.
- the manufactured articles have manufacturing tolerances, so that different products or articles can emit different noise signals due to production tolerances.
- aging or load influences as well as manufacturing tolerances also affect the room characteristics of the environment in which the device or the system is located emitted by the device or the system noise signals or acoustic emissions.
- a plurality of recordings are made on a manufactured prototype in order to generate a model of the respective noise-generating machine or system.
- a large number of recordings in particular sound recordings, are carried out, which take various factors into account. For example, acoustic sound signals emitted by a prototype are recorded at different times under different weather conditions and under different loads. In order to take into account various factors, a large number of images must be taken.
- the model generation based on the prototype is the more complex the more different variants of the product are available.
- a statistical model or a physical model of the object or product is generated, which can then be used to classify a noise signal which is emitted by a manufactured product after startup during operation. Due to the classification of the noise signal, an operating state of a product can be monitored so that occurring fault conditions and thus the need for maintenance work are detected.
- a disadvantage of the conventional procedure is that the sound signals or acoustic emissions emitted by a manufactured product also differ from or differ from the noise signals of the prototype in a normal or error-free state.
- the manufactured product which may be, for example, a machine, device or equipment, is in a different environment than the prototype when recording the training data.
- the useful signal that is the acoustic emissions of the examining object, ambient noise superimposed as interference signal.
- the training data may be recorded on a prototype in a low reverberation room while the subject being examined, such as a manufacturing machine, is located in a factory building that strongly reflects acoustic signals.
- the product to be investigated or the system to be investigated may have a different acoustic emission spectrum than the prototype.
- the invention provides a method for detecting a state of a sound-generating object to be examined, wherein a basic statistical classification model of acoustic features generated for at least one reference object is automatically adapted based on acoustic characteristics of a noise generated by the object to be examined and based on the adapted statistical Classification model is a classification of the state of the noise-generating object to be examined.
- the sound generating object may be any machine or equipment, such as a chemical plant.
- the reference object is formed by a prototype of the object to be examined.
- the noise-generating object to be examined has at least one motor-driven noise-generating module.
- the adaptation of the statistical classification model takes place during the commissioning of the object to be examined, at regular maintenance intervals or when an acoustic environment of the object to be examined changes.
- the basic statistical classification model of acoustic features is generated on the basis of a plurality of acoustic sound recordings that are made on the reference subject under different conditions or at different times.
- the generated basic statistical classification model is stored in a memory.
- the stored basic classification model is automatically adapted on the basis of a number of acoustic sound recordings which are made on the object to be examined generating noise and buffered as an adapted statistical classification model.
- the acoustic sound recordings are made by means of acoustic Tonierêt that detect airborne or solid-borne sound.
- the statistical classification model is adapted for the entire noise-generating article or for modules of the sound-generating article.
- the acoustic sound pickups are attached to the sound-generating object to be examined or are guided past the sound-generating object to be examined.
- an adapted statistical classification model is calculated in each case for differently adjustable positions of the acoustic sound pickup past the object to be examined.
- an adapted statistical classification model is calculated for the adjustable positions of the acoustic acoustic pickup.
- the adjustable positions of the passing acoustic sound pickup correspond to the spatial arrangement of different modules of the sound-generating object to be examined.
- the invention further provides an apparatus for detecting a condition of a sound-generating object to be examined, wherein a basic statistical classification model of acoustic features generated for at least one reference object is automatically adapted by a data processing unit based on acoustic characteristics of a sound generated by the object to be examined the data processing unit classifies the state of the sound-generating object to be examined on the basis of the adapted statistical classification model.
- the latter has at least one acoustic sound pickup for generating a body or airborne sound generated by the sound-generating object to be examined.
- FIG. 1 shows a flowchart for illustrating a possible embodiment of the method according to the invention
- Figure 2 is another diagram illustrating the procedure in the inventive method
- Figure 3 is a block diagram of a possible embodiment of the device according to the invention for detecting a state of a sound-generating object to be examined;
- FIGS. 4A, 4B show further embodiments of the device according to the invention for detecting a state of a sound-generating object to be examined.
- a step S1 sound recordings are first made by means of sound pickups or sensors on a reference object of the sound-generating object to be examined.
- the reference item may be a prototype of the item.
- the object is for example a machine or a device which generates noise during operation.
- the noise signal may be an airborne sound signal or a structure-borne noise signal.
- the object under investigation can also be a plant, for example a chemical plant, which causes noise during a chemical production process, for example inside tank equipment or pipes.
- the noise or vibration signal is recorded or recorded with sound pickups.
- These Tonaufsacrificing have microphones or other vibration sensors, in particular acceleration sensors on.
- the Tonauflutter commentary or vibration sensors convert the noise or vibration signal into an electrical signal.
- the sound recordings on the reference object or on the prototype are preferably made at different times under different environmental conditions.
- step S2 features m of the sound signal or vibration signal are extracted on the basis of the sound recordings.
- Possible features are, for example, the amplitude or the volume of the sound or noise signal.
- Further possible features are spectral features of the noise signal and their temporal modulations.
- features of the modulation spectrum can be used.
- a feature vector M can be determined in each case. Typically, 10-500 features m of the noise signal are considered.
- the number N of shots on the reference object is set depending on the object under consideration. Usually, N> 100 sound recordings of the reference subject are evaluated.
- a basic classification model is generated in step S2 by means of an algorithm. The model generation can be carried out, for example, by means of Expectation Maximization (EM) or GMM, by means of a Support Vector Machine (SVM), by means of Seif Organizing Maps (SOM) or by means of neural networks.
- EM Expectation Maximization
- SVM Support Vector Machine
- SOM Seif Organizing Maps
- acoustic airborne or structure-borne noise signals are recorded or recorded by means of sound pickups on the object to be examined.
- the im Step S2 generated basic statistical classification model of acoustic features m is automatically adapted in step S3 on the basis of acoustic characteristics m, the noise signal recorded in step S3.
- the basic classification model has at least one statistical parameter P for each feature m. These statistical parameters P include, for example, an average ⁇ or a variance ⁇ 2 .
- the features m can be based on a Gaussian distribution, wherein for each feature m based on the sound recordings of the reference object an average value ⁇ and a variance ⁇ 2 of the distribution is given.
- step S1 sound recordings are made at different reference objects or prototypes in step S1 and an average value ⁇ is calculated with respect to each feature m.
- the feature m is the volume of the detected sound signal
- the mean ⁇ of the volume and a variance ⁇ 2 of the volume are given as statistical parameters P for the feature of the volume in the basic classification model.
- An adaptation of the basic classification model then takes place in step S4 on the basis of the sound recording made to the object to be examined in step S3, in that the statistical parameters P are adapted accordingly for each feature m of the basic classification model.
- Sound recording on the object to be examined in step S3 and the adaptation of the basic classification model in step S4 can be carried out in a possible embodiment of the method according to the invention during the commissioning of the object to be examined.
- the sound recording in step S3 and the adaptation in step S4 takes place at regular maintenance intervals, for example once a day.
- the sound recording in step S3 and the adaptation of the basic classification model takes place in step S4 in a change of an acoustic environment of the object to be examined, for example, when the object is placed in another room.
- the sound recording and the adaptation can be made on demand, for example after carrying out a maintenance work on the object to be examined or if a user considers an investigation to be necessary.
- a classification of the state of the sound-generating object to be examined is carried out on the basis of the adapted statistical classification model.
- This state classification can be done on the basis of all or some of the features m of the adapted basic classification model. For example, a sound-producing object to be examined may be classified as faulty if the feature volume exceeds an adjustable threshold.
- step S6 the classified state of the object to be examined is output and, if necessary, necessary measures are then initiated.
- FIG. 2 shows a diagram for clarifying a possible embodiment of the method according to the invention for recognizing a condition of a sound-generating object to be examined.
- the article shown in Figure 2 is a machine that provides acoustic data.
- Sound recordings are performed on n prototypes of the machine and a universal basic statistical classification model is generated on the basis of the training data.
- This universal basic classification model is stored in a memory or a database.
- This general classification model preferably has the essential properties of a normal state and / or possible errors or maintenance conditions of the object or the machine.
- This is followed by an adaptation of the universal basic classification model for the various manufactured machines. In the example shown in FIG. 2, various machines in the field are examined.
- Each machine provides an acoustic noise signal or acoustic data, which are used for individual adaptation of the stored universal classification model.
- an associated adaptive classification model is created for each machine, which can be temporarily stored in a memory.
- a classification algorithm a classification of the state of the respective noise-generating machine to be examined is then carried out on the basis of the adapted statistical classification model in order to detect an operating state or to generate an error message.
- the considerable effort required to generate a classification model is carried out only once on the basis of at least one prototype.
- the subsequent adaptation of the basic statistical classification model formed can be done with a relatively low computational effort in a short time or in real time.
- the adaptation of the stored basic classification model for the respective object to be examined can be automated with little computational effort under changed environmental conditions in the field, without a corresponding Expert or specialist must be on site.
- comparatively few additional acoustic data or sound recordings on the respective machine or on the respective object are required for the adaptation of the statistical basic classification model.
- the method according to the invention provides for a two-stage approach, namely first the formation of a universal statistical basic classification model and a subsequent adaptation of this classification model for the respective object to be examined in the field.
- FIG. 3 shows an exemplary embodiment of a device 1 according to the invention for detecting a state of a sound-generating object to be examined.
- the object 2 to be examined is, for example, a device or a machine.
- the device 1 has at least one pickup 3 for detecting a body or airborne sound signal generated by the sound-generating object 2 to be examined.
- the pickup 3 detects the emitted by the object 2 vibration signal.
- the pickup 3 is, for example, an acoustic pickup for detecting an airborne sound signal, that is, a microphone.
- the pickup 3 is mounted directly on the object 2 and detects structure-borne noise or vibrations of the object 2.
- a pickup 3 may be, for example, an acceleration sensor.
- the pickup 3 converts the sound or vibration signal into an electrical signal and outputs it via a line 4 to a data processing unit 5.
- the data processing unit 5 is, for example, a microprocessor in which a program is executed.
- the data processing unit 5 is connected via lines 6 to a memory 7 in which the generated basic classification model is stored.
- the data processing unit 5 receives the basic classification model via an interface.
- the data processing unit 5 adapts the basic classification model of acoustic characteristics on the basis of acoustic characteristics m of a noise generated by the object 2 to be examined.
- the state of the sound-generating object 2 to be examined is classified by the data processing unit 5 by means of a classification algorithm. This condition is reported via output lines 8.
- the pickup 3 and the data processing unit 5 are integrated in one component. This device can be attached to any object 2.
- FIGS. 4A, 4B show further embodiments of the device 1 according to the invention for detecting a state of a sound-generating object 2 to be examined.
- the noise-generating object 2 is a machine or a device which has a multiplicity of identical machine modules 2-1, 2-2 to 2-k.
- An example of such a machine 2 is a letter sorting system which has a plurality of segments or sorting Ehern same design has.
- the various segments or modules of Briefsortierstrom 2 may be exposed to a different load and therefore have a different wear.
- Such BriefSortieranlagen can be in different acoustic environments.
- an associated sensor 3 is attached to each module of the machine 2 and is provided for monitoring the operating state of the respective module on the basis of the noise signal emitted by the module.
- the data processing unit 5 then carries out the steps S3-S6 of the sequence shown in FIG. 1 for each module 2-i of the machine 2.
- a pickup 3 is not provided for each module 2-i of the machine 2, but instead a pickup, for example a pickup, is guided linearly past the machine 2 and outputs the detected noise signal, for example via a pickup 3 Air interface to the data processing unit 5 from.
- a pickup for example a pickup
- an adapted statistical classification model can be calculated by the data processing unit 5 for the different positions of the acoustic sound pickup 3 guided past the object 2 to be examined. For example, in a letter sorting system, the distance ⁇ between the various segments or sorting compartments is known.
- the various adjustable positions of the passed acoustic sound pickup 3 correspond to the spatial arrangement of the various modules 2-i of the sound-generating letter sorter 2 to be examined.
- an adapted classification model for the entire BriefSortierstrom 2 is created in the field of a universal classification model of a prototype.
- the acoustic monitoring of the various modules 2-i can be done by vibration sensors or microphones.
- the vibration sensors are mounted directly on the housing and detect a structure-borne noise.
- the microphones can have a specific directional characteristic in one possible embodiment.
- the distance of the microphones or the receiver 3 to the object 2 to be examined may vary between a few millimeters and several meters.
- the acoustic monitoring takes place by means of only one sound pickup 3, which is automatically or manually guided past the location of a fault on the letter sorting system 2.
- a plurality of acoustic sensors 3 are provided, which each observe a part of the system or of the object 2 to be examined.
- the adaptation can be carried out by moving the acoustic sensor 3 to different characteristic positions of the system and recording acoustic data there.
- the recorded acoustic data is used to generate the adapted statistical classification model.
- a sound pickup 3 can record data and calculate a separate adapted model for itself.
- This adapted model can be stored locally or alternatively centrally in the sensor or acoustic pickup.
- the sound pickup 3 can be moved to different positions of the letter sorter system 2 and record there the acoustic data on site.
- the sound pickup 3 can calculate an adapted classification model for the respective position of the sound pickup 3 or from a number of comparable positions For example, for machine segments or modules of the same type, calculate adaptive classification models.
- BriefSortierstrom 2 is done, the transducer or 3 is brought into a "Classification" operating state and an error detection activated. If necessary, the adaptation can be repeated at any time, for example when components or entire modules are replaced or when room acoustics change.
- a model already adapted during the commissioning of the letter sorting installation 2 can be used, depending on whether the acoustic spatial situation has changed or not.
- the method according to the invention for detecting a state of a sound-generating object 2 to be examined is suitable for any objects, such as, for example, for monitoring motors, pumps, automation systems or chemical systems.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Combustion & Propulsion (AREA)
- Chemical & Material Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Soundproofing, Sound Blocking, And Sound Damping (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Control Of Electric Motors In General (AREA)
Abstract
Description
Claims
Priority Applications (7)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
RU2010148369/28A RU2494364C2 (ru) | 2008-04-29 | 2009-04-29 | Способ и устройство для распознавания состояния исследуемой создающей шумы машины |
BRPI0911440A BRPI0911440A2 (pt) | 2008-04-29 | 2009-04-29 | método e dispositivo para reconhecer um estado de uma máquina geradora de ruído a ser investigada |
MX2010011772A MX2010011772A (es) | 2008-04-29 | 2009-04-29 | Metodo y dispositivo para reconocer un estado de una maquina a analizar que genera ruido. |
CN2009801147528A CN102016535B (zh) | 2008-04-29 | 2009-04-29 | 用于对产生噪声的待检机器进行状态识别的方法及装置 |
US12/736,675 US9714884B2 (en) | 2008-04-29 | 2009-04-29 | Method and device for recognizing state of noise-generating machine to be investigated |
AT09738167T ATE535794T1 (de) | 2008-04-29 | 2009-04-29 | Verfahren und vorrichtung zum erkennen eines zustandes einer zu untersuchenden geräuscherzeugenden maschine |
EP09738167A EP2271908B1 (de) | 2008-04-29 | 2009-04-29 | Verfahren und vorrichtung zum erkennen eines zustandes einer zu untersuchenden geräuscherzeugenden maschine |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE102008021362A DE102008021362B3 (de) | 2008-04-29 | 2008-04-29 | Verfahren und Vorrichtung zum Erkennen eines Zustandes einer zu untersuchenden geräuscherzeugenden Maschine |
DE102008021362.4 | 2008-04-29 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2009133123A1 true WO2009133123A1 (de) | 2009-11-05 |
WO2009133123A4 WO2009133123A4 (de) | 2009-12-17 |
Family
ID=40691029
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2009/055165 WO2009133123A1 (de) | 2008-04-29 | 2009-04-29 | Verfahren und vorrichtung zum erkennen eines zustandes einer zu untersuchenden geräuscherzeugenden maschine |
Country Status (9)
Country | Link |
---|---|
US (1) | US9714884B2 (de) |
EP (1) | EP2271908B1 (de) |
CN (1) | CN102016535B (de) |
AT (1) | ATE535794T1 (de) |
BR (1) | BRPI0911440A2 (de) |
DE (1) | DE102008021362B3 (de) |
MX (1) | MX2010011772A (de) |
RU (1) | RU2494364C2 (de) |
WO (1) | WO2009133123A1 (de) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102811386A (zh) * | 2011-06-01 | 2012-12-05 | 中兴通讯股份有限公司 | 一种录音装置、媒体服务器及录音方法、系统 |
CN113489514A (zh) * | 2021-07-05 | 2021-10-08 | 国网湖南省电力有限公司 | 基于自组织映射神经网络的电力线通信噪声识别方法及装置 |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102008021362B3 (de) | 2008-04-29 | 2009-07-02 | Siemens Aktiengesellschaft | Verfahren und Vorrichtung zum Erkennen eines Zustandes einer zu untersuchenden geräuscherzeugenden Maschine |
DE102008046286B4 (de) | 2008-09-08 | 2010-04-29 | Siemens Aktiengesellschaft | Ortungseinheit zur Anbringung an einem bewegbaren Objekt |
DE102008046285A1 (de) | 2008-09-08 | 2010-03-25 | Siemens Aktiengesellschaft | Vorrichtung zur Analyse von Atmungsgeräuschen |
DE102014207784A1 (de) * | 2014-04-25 | 2015-10-29 | Robert Bosch Gmbh | Verfahren zur Symptomerkennung von Fehlfunktionen eines Elektrowerkzeugs |
CN104111109B (zh) * | 2014-07-21 | 2015-09-02 | 石家庄铁道大学 | 一种基于不同阶次统计量及支持向量机的机械振动状态识别方法 |
CN108597057A (zh) * | 2018-04-28 | 2018-09-28 | 济南浪潮高新科技投资发展有限公司 | 一种基于噪音深度学习的无人机故障预测诊断系统及方法 |
US10773665B2 (en) * | 2018-10-16 | 2020-09-15 | Cnh Industrial America Llc | System and method for detecting a damage condition associated with an agricultural machine |
CN110686899B (zh) * | 2019-09-21 | 2021-01-29 | 天津大学 | 一种内燃机噪声源识别方法 |
CN113696839B (zh) * | 2021-10-29 | 2021-12-28 | 南京易砼科技有限公司 | 一种混凝土泵送车辆的泵送状态检测方法和装置 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1564688A1 (de) * | 2004-02-12 | 2005-08-17 | Deere & Company | Verfahren und Überwachungssystem zur Überwachung des Zustands von Arbeitsmaschinen |
DE102004012911A1 (de) * | 2004-03-17 | 2005-10-06 | Daimlerchrysler Ag | Verfahren zur Bewertung eines Betriebsgeräusches eines Kraftfahrzeugmotors |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE29611558U1 (de) * | 1996-07-05 | 1997-08-07 | Siemens Ag | Vorrichtung zur Erfassung von analogen Meßsignalen für die akustische Diagnose von Prüflingen |
EP1088211B1 (de) * | 1998-04-17 | 2003-06-25 | Siemens Aktiengesellschaft | Akustisches diagnosesystem und -verfahren |
DE19914533A1 (de) * | 1999-03-31 | 2000-11-02 | Bayerische Motoren Werke Ag | Vibrationsakustisches Diagnosesystem |
US20010044719A1 (en) * | 1999-07-02 | 2001-11-22 | Mitsubishi Electric Research Laboratories, Inc. | Method and system for recognizing, indexing, and searching acoustic signals |
DE10100522B4 (de) * | 2001-01-08 | 2013-03-28 | Deere & Company | Überwachungseinrichtung zur Überwachung der Funktion einer Arbeitsmaschine |
CN1320372C (zh) * | 2004-11-25 | 2007-06-06 | 上海交通大学 | 基于小波域的水声噪声的检测与识别方法 |
US8744844B2 (en) * | 2007-07-06 | 2014-06-03 | Audience, Inc. | System and method for adaptive intelligent noise suppression |
DE102008021362B3 (de) | 2008-04-29 | 2009-07-02 | Siemens Aktiengesellschaft | Verfahren und Vorrichtung zum Erkennen eines Zustandes einer zu untersuchenden geräuscherzeugenden Maschine |
EP2133866B1 (de) * | 2008-06-13 | 2016-02-17 | Harman Becker Automotive Systems GmbH | Adaptives Geräuschdämpfungssystem |
-
2008
- 2008-04-29 DE DE102008021362A patent/DE102008021362B3/de not_active Expired - Fee Related
-
2009
- 2009-04-29 MX MX2010011772A patent/MX2010011772A/es active IP Right Grant
- 2009-04-29 BR BRPI0911440A patent/BRPI0911440A2/pt not_active IP Right Cessation
- 2009-04-29 EP EP09738167A patent/EP2271908B1/de not_active Not-in-force
- 2009-04-29 RU RU2010148369/28A patent/RU2494364C2/ru not_active IP Right Cessation
- 2009-04-29 US US12/736,675 patent/US9714884B2/en not_active Expired - Fee Related
- 2009-04-29 CN CN2009801147528A patent/CN102016535B/zh not_active Expired - Fee Related
- 2009-04-29 AT AT09738167T patent/ATE535794T1/de active
- 2009-04-29 WO PCT/EP2009/055165 patent/WO2009133123A1/de active Application Filing
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1564688A1 (de) * | 2004-02-12 | 2005-08-17 | Deere & Company | Verfahren und Überwachungssystem zur Überwachung des Zustands von Arbeitsmaschinen |
DE102004012911A1 (de) * | 2004-03-17 | 2005-10-06 | Daimlerchrysler Ag | Verfahren zur Bewertung eines Betriebsgeräusches eines Kraftfahrzeugmotors |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102811386A (zh) * | 2011-06-01 | 2012-12-05 | 中兴通讯股份有限公司 | 一种录音装置、媒体服务器及录音方法、系统 |
CN113489514A (zh) * | 2021-07-05 | 2021-10-08 | 国网湖南省电力有限公司 | 基于自组织映射神经网络的电力线通信噪声识别方法及装置 |
CN113489514B (zh) * | 2021-07-05 | 2022-07-26 | 国网湖南省电力有限公司 | 基于自组织映射神经网络的电力线通信噪声识别方法及装置 |
Also Published As
Publication number | Publication date |
---|---|
ATE535794T1 (de) | 2011-12-15 |
RU2010148369A (ru) | 2012-06-10 |
CN102016535A (zh) | 2011-04-13 |
EP2271908A1 (de) | 2011-01-12 |
CN102016535B (zh) | 2012-10-03 |
EP2271908B1 (de) | 2011-11-30 |
MX2010011772A (es) | 2010-12-21 |
US20110047107A1 (en) | 2011-02-24 |
DE102008021362B3 (de) | 2009-07-02 |
RU2494364C2 (ru) | 2013-09-27 |
WO2009133123A4 (de) | 2009-12-17 |
US9714884B2 (en) | 2017-07-25 |
BRPI0911440A2 (pt) | 2015-10-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2271908B1 (de) | Verfahren und vorrichtung zum erkennen eines zustandes einer zu untersuchenden geräuscherzeugenden maschine | |
EP2277039B1 (de) | Verfahren und vorrichtung zur klassifikation von schallerzeugenden prozessen | |
DE69732569T2 (de) | System zur erkennung von störungen von elektromotoren | |
DE102017009471A1 (de) | Humankollaboratives robotersystem mit verbesserter genauigkeit der erfassung einer äusseren kraft durch maschinelles lernen | |
DE102019113830A1 (de) | Informationsverarbeitungsverfahren, Informationsverarbeitungsgerät und Programm | |
DE102019115237A1 (de) | Verfahren und vorrichtung zum überwachen eines fahrzeugseitigen maschinenlagers | |
DE102018133533A1 (de) | Überprüfungssystem und Verfahren zum Überprüfen von Arbeitsabläufen | |
DE102014012184A1 (de) | Vorrichtung und Verfahren zur automatischen Erkennung und Klassifizierung von akustischen Signalen in einem Überwachungsbereich | |
DE102019100722A1 (de) | Erkennung von Spitzen und Fehlern in Schwingungstrendddaten | |
DE10320809A1 (de) | Verfahren zur Erkennung und Überwachung der Bewegung bei Fahrzeugen | |
DE10235416A1 (de) | Diagnoseverfahren für Kraftfahrzeuge | |
DE10062606A1 (de) | Verfahren und Einrichtung zum Überwachen des mechanischen Zustands von elektrisch angetriebenen Fahrzeugen im regulären Fahrbetrieb | |
DE102022203238A1 (de) | Verfahren zur Betriebspunktklassifizierung durch Schallsignale | |
DE102008060194B4 (de) | Verfahren und Kraftfahrzeug für ein Fahrzeugflotten-Qualifikationsmanagement | |
EP3757698A1 (de) | Verfahren und vorrichtung zur bewertung und auswahl von signal-vergleichsmetriken | |
DE102022200285B3 (de) | Verfahren und Vorrichtung zum Bereitstellen eines datenbasierten Systemmodells und zum Überprüfen eines Trainingszustands des Systemmodells | |
EP3466792A1 (de) | Verfahren und system zum akustischen analysieren eines stellwerkraums | |
DE102019220310B3 (de) | Verfahren zur Erfassung von statisch mit einem Kraftfahrzeug in Kontakt stehenden Objekten und Kraftfahrzeug | |
DE102021209414A1 (de) | Diagnose eines Kraftfahrzeuges in Fahrzuständen durch Audioanalyse | |
DE102021102712A1 (de) | Verfahren und System zur akustischen Diagnose eines Kraftfahrzeugs anhand von Betriebsgeräuschen | |
EP1465791A1 (de) | Verfahren zum bewerten eines einbauorts einer beschleunigungssensor-baugruppe in einem fahrzeug | |
DE102020209255A1 (de) | Pneumatisches System und Verfahren | |
DE102022208868A1 (de) | Maschinelles Lernsystem zur optischen Inspektion | |
DE102013013551A1 (de) | Verfahren zur Erkennung signifikant veränderter Signale vor einem Hintergrund dominanter Störungen | |
DE102021210667A1 (de) | Vorverarbeitung eines Eingangssignals |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200980114752.8 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09738167 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2009738167 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 7207/DELNP/2010 Country of ref document: IN |
|
WWE | Wipo information: entry into national phase |
Ref document number: MX/A/2010/011772 Country of ref document: MX |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12736675 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2010148369 Country of ref document: RU |
|
ENP | Entry into the national phase |
Ref document number: PI0911440 Country of ref document: BR Kind code of ref document: A2 Effective date: 20101028 |